Local Patch AutoAugment with Multi-Agent Collaboration
arXiv:2103.11099
Abstract
Data augmentation (DA) plays a critical role in improving the generalization of deep learning models. Recent works on automatically searching for DA policies from data have achieved great success. However, existing automated DA methods generally perform the search at the image level, which limits the exploration of diversity in local regions. In this paper, we propose a more fine-grained automated DA approach, dubbed Patch AutoAugment, to divide an image into a grid of patches and search for the joint optimal augmentation policies for the patches. We formulate it as a multi-agent reinforcement learning (MARL) problem, where each agent learns an augmentation policy for each patch based on its content together with the semantics of the whole image. The agents cooperate with each other to achieve the optimal augmentation effect of the entire image by sharing a team reward. We show the effectiveness of our method on multiple benchmark datasets of image classification and fine-grained image recognition (e.g., CIFAR-10, CIFAR-100, ImageNet, CUB-200-2011, Stanford Cars and FGVC-Aircraft). Extensive experiments demonstrate that our method outperforms the state-of-the-art DA methods while requiring fewer computational resources.
References in corpus (12)
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- mixup: Beyond Empirical Risk Minimization
- Improved Regularization of Convolutional Neural Networks with Cutout
- Wide Residual Networks
- Fine-Grained Visual Classification of Aircraft
- AutoAugment: Learning Augmentation Policies from Data
- Counterfactual Multi-Agent Policy Gradients
- Shake-Shake regularization
- Population Based Augmentation: Efficient Learning of Augmentation Policy Schedules
- A Bayesian Data Augmentation Approach for Learning Deep Models
- Affinity and Diversity: Quantifying Mechanisms of Data Augmentation
- Boundary-aware Supervoxel-level Iteratively Refined Interactive 3D Image Segmentation with Multi-agent Reinforcement Learning